EUVIMEDEUVIMED entdecken
Crossref

Data Geometry and Extreme Value Distribution

Mamadou Cisse, Aliou Diop, Souleymane Bognini, Nonvikan Karl-Augustt ALAHASSA

Journal of Mathematics and Statistics Studies · 2021

DOI: 10.32996/jmss.2021.2.2.2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Mathematics and Statistics Studies, 2021. Verfasst von Mamadou Cisse und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In extreme values theory, there exist two approaches about data treatment: block maxima and peaks-over-threshold (POT) methods, which take in account data over a fixed value. But, those approaches are limited. We show that if a certain geometry is modeled with stochastic graphs, probabilities computed with Generalized Extreme Value (GEV) Distribution can be deflated. In other words, taking data geometry in account change extremes distribution. Otherwise, it appears that if the density characterizing the states space of data system is uniform, and if the quantile studied is positive, then the Weibull distribution is insensitive to data geometry, when it is an area attraction, and the Fréchet distribution becomes the less inflationary.

Zitieren

APA

Cisse, M., Diop, A., Bognini, S., & ALAHASSA, N. K. (2021). Data Geometry and Extreme Value Distribution. Journal of Mathematics and Statistics Studies. https://doi.org/10.32996/jmss.2021.2.2.2

Vancouver

Cisse M, Diop A, Bognini S, ALAHASSA NK. Data Geometry and Extreme Value Distribution. Journal of Mathematics and Statistics Studies. 2021. doi:10.32996/jmss.2021.2.2.2

BibTeX
@article{Cisse2021,
  author = {Cisse, Mamadou and Diop, Aliou and Bognini, Souleymane and ALAHASSA, Nonvikan Karl-Augustt},
  title = {Data Geometry and Extreme Value Distribution},
  journal = {Journal of Mathematics and Statistics Studies},
  year = {2021},
  doi = {10.32996/jmss.2021.2.2.2}
}
Kostenlos starten →